METHOD FOR LOCATING A TARGET USING A MULTI-QUADRANT DETECTOR

Description

TECHNICAL FIELD

[0001] The invention relates to the field of target detection using a multi-quadrant detector, the target preferably being illuminated. The invention finds particular application in the field of semi-active laser guidance of an object towards the target for the automatic guidance of air / surface or surface / surface weapons. STATE OF THE ART

[0002] A target location device comprising a four-quadrant photodetector uses a flux diffused by the target to be detected to measure the direction of the target in its aiming frame. Such location makes it possible in particular to track a moving target such as an aircraft, machine, missile, vehicle, tank, etc. Such a device can be mounted on a guidance system, for example a self-guided missile, or on a firing station making it possible to track a beacon carried by the missile and a target.

[0003] The difference between the line of sight and the direction of the target is the deviation or angular deviation or angular misalignment. This deviation is usually, in the case of guidance, either brought back to zero or kept constant so that the guidance system including the location device is correctly guided towards the target.

[0004] The target can naturally emit light. However, in semi-active laser guidance, a high-power, low-divergence laser beam emitted by a designator marks the target, and the laser flux scattered (backscattered) by the target is then used to measure the deviation. The designator is, for example, integrated into the sight or carried by a third-party, cooperative device.

[0005] In particular, we know of two categories of guidance systems or sensors which exploit such a laser flux diffused by the target to be located.

[0006] Laser locators, or LSTs, from the English, Laser-Spot-Tracker, are devices integrated into a sight whose function is to ensure that the laser beam emitted by the designator is correctly positioned on the target. An LST is more precisely a deviation meter that determines the angular direction of the laser flux diffused by the target and this direction is compared to the direction of the target observed in an imager harmonized with the LST.

[0007] Semi-Active Laser Seekers or E-SALs, referred to as E-SALs, are integrated into weapons and provide them with the ability to accurately guide themselves to the laser spot positioned on the target, which can be fixed, movable or mobile. Document US2011073704A1 discloses a seeker or deviator of this type.

[0008] The main functions of these systems are: the detection of a laser flux reflected by the pointed target and associated with the designation laser; the tracking of this target from the reflection of the laser flux maintained positioned on it; the provision of a deviation measurement of the target in the reference frame of the LST sensor or the E-SAL sensor.

[0009] For localization and guidance, a common solution is to use a multi-quadrant photodetector, in particular a 4-quadrant type photodetector, whose photodetector surface is in the shape of a disc or square cut into four equal and independent photodetection surfaces. This photodetector is associated with a flux-collecting optic allowing a more or less localized energy distribution, typically a homothetic disc of the illumination in the entrance pupil of the optic, on the active surface of the photodetector. The position of this energy distribution depends on the angular direction of origin of the incident flux.

[0010] The energy levels collected by each of the quadrants therefore vary according to this direction and this is what makes it possible to determine the deviation.

[0011] Let's take a four-quadrant photodetector as an example. To determine the deviation, the operations to be carried out with a 4-quadrant detector consist of two comparisons of the signals received on the paired sectors: on the one hand, the sum of the 2 quadrants at the top vs. the sum of the 2 quadrants at the bottom and on the other hand, the sum of the 2 quadrants on the left vs. the sum of the 2 quadrants on the right. Two weighings of the detected signal are therefore carried out along the vertical and horizontal axes. Then a tabulation previously carried out in the factory is applied and gives the true angular positions in elevation and bearing of the spot relative to the detector's reference frame, in correspondence with the top-bottom and left-right weighing values.

[0012] This conversion step between weighing and deviation measurement assumes that the energy distribution on the 4-quadrant detector is identical in the means implemented in the factory and in the context of use.

[0013] But this is not always the case.

[0014] The accuracy of deviation measurement is sometimes degraded, particularly in the presence of atmospheric turbulence or strong light.

[0015] Indeed, the laser signals generated by the four quadrants of the photodetector, corresponding to the illumination of the sensor coming from the designation of the target, are likely to be disturbed: a) parasitic sources emit a photonic flux partially included in the spectral band of the optical filter of the sensor and with a modulation which is not totally rejected by the electronic filtering adapted to the temporal modulation of the designation laser; these stimuli are added to the laser signal and can potentially modify the response of the system with respect to the useful laser signal; b) the scene environment influences the propagation of the wavefront of the designator laser towards the target then from the target towards the sensor; the turbulence present on these paths generate scintillation-type effects which result in modulations of illumination inside the optical pupil of the sensor.

[0016] Type a) disturbances result from the problem described below.

[0017] The electronic chains, both analog and digital, have the functions of filtering the signal captured by each channel or quadrant of the 4-quadrant photodetector, for the purposes of detection and recognition of time coding; and of developing the information necessary to measure the 4 energy portions of the laser pulses, for the purposes of target localization. The elements of the electronic chains must ensure good restitution of laser deviations, including in cases of multi-echoes with 2 temporally close pulses.

[0018] These requirements impose a compromise, during design, between the sensitivity of the sensor, its instantaneous measurement dynamics, the effectiveness of parasite filtering and the fidelity in the restitution of the 4 energy information and therefore in the deviation accuracy.

[0019] Distortion or saturation of one of the 4 useful signals may occasionally occur in the presence of intense parasitic modulations.

[0020] The signals from the four quadrants and used to perform the calculations for the two weighings are then biased by the amputation of the laser signal on the distorted or saturated quadrant(s).

[0021] Concerning type b) disturbances, the problem is as follows.

[0022] The variations in the optical index of the air along the laser path result in inhomogeneous illumination on the optical pupil of the sensor, and these inhomogeneities vary from one laser pulse reception to the next. This non-uniformity, which varies over time, is reflected by modulations in the energy distribution on the photodetector and disrupts the two weighings.

[0023] For small disturbances, the deviation noise is aggravated, but without jeopardizing the success of the mission.

[0024] For strong disturbances, the dispersion of the delivered deviation measurements makes localization and guidance unsuccessful. STATEMENT OF THE INVENTION

[0025] The invention proposes to overcome at least one of these drawbacks and in particular proposes to diagnose the two types of degraded operation: (a) saturation of the electronic chain in the presence of strong background modulation and (b) inhomogeneity of pupil illumination in the presence of turbulence.

[0026] To this end, the invention relates to a method for locating a target by means of a detection device comprising a photodetector comprising NxM quadrants positioned relative to a line of sight of the detection device, the photodetector being configured to detect optical signals originating from the target to be detected, said method being implemented in a processing unit of said device and comprising the following steps, the target diffusing an optical signal: acquisition of the signals detected by each quadrant corresponding to the signals broadcast by the target towards the quadrants; processing of the acquired signals to determine a deviation in elevation and a deviation in bearing of the direction of the target relative to the line of sight of the device; processing of the acquired signals to deduce therefrom indicators of distribution of the signals detected by the quadrants; processing of the deviations in elevation and bearing determined to deduce therefrom indicators of the theoretical distribution of the signals in the quadrants for these deviations; the indicators of distribution of the signal in the quadrants being obtained by means of the theoretical and measured weight values, said weights being a function of the signal levels detected by the quadrants. verification of the consistency of the indicators of the theoretical distribution with the indicators of the distribution resulting from the acquired signals so as to determine whether the measured deviation is valid.

[0027] The invention is advantageously supplemented by the following characteristics, taken alone or in any of their technically possible combinations: the indicators of the theoretical distribution of the signals in the quadrants is obtained by interpolation, preferably bilinear, of the deviations with a mapping of the indicators of the theoretical distribution of the signals as a function of several theoretical deviations; the photodetector comprises quadrants distributed along two perpendicular axes called site Z and bearing Y axes, the indicators of distribution of the signal in the quadrants being obtained by means of the theoretical and measured weighing values, said weighings being a function of the signal levels detected by quadrants;the NxM quadrants of the photodetector being distributed along two axes perpendicular to each other, called the Z site axis and the Y bearing axis, the quadrants detecting a signal defining a sub-matrix included in the NxM matrix, of size NixMi quadrants with Ni less than or equal to N, and Mi less than or equal to M, and detecting signals, the signal distribution indicators being obtained by comparing the signals detected on adjacent quadrants taken along the rows of the NixMi matrix on the one hand and taken along the columns of the NixMi matrix on the other hand; the consistency check consists of calculating a comparison criterion between a first metric based on the theoretical distribution indicators and a second metric based on the distribution indicators of the acquired signals, the criterion being compared to a threshold to validate the deviation measurements;the comparison criterion is defined by an unsigned difference between the first and second metrics, the deviation being valid if this criterion is less than a threshold and invalid if this criterion is greater than or equal to said threshold; the threshold is adjusted according to the signal-to-noise ratio measured on each quadrant and / or the determined deviation; the photodetector comprises four quadrants A, B, C, D, quadrants A and B being above the site axis Z, quadrants C and D being below the Z axis, quadrant C being below quadrant A, quadrant D being below quadrant B, the criteria for distributing the signals on the quadrants being defined as follows: Y 1 =(BA) / (A+B), weighing comparing the two signals from the top; Y 2 =(DC) / (C+D), weighing comparing the two bottom signals Y=(B+DAC) / (A+B+C+D) weighing comparing the signals on either side of the vertical axis;Z 1 =(AC) / (A+C), weighing comparing the two signals on the left Z 2 =(BD) / (B+D), weighing comparing the two signals on the right Z=(A+BCD) / (A+B+C+D) weighing comparing the signals on either side of the horizontal axis. the metrics as a function of the theoretical or measured distribution criteria is defined by max([abs(Y 1 -Y 2 ) ; abs(Z 1 -Z 2 ) ; abs(Y 1 -Y) ; abs(Z 1 -Z) ; abs(Y 2 -Y), abs(Z 2 -Z)]) with max the maximum value and abs the absolute value. ;

[0028] The invention also relates to a device for detecting a target comprising a multi-quadrant photodetector and a processing unit configured to implement a method according to the invention.

[0029] The invention also relates to a system for guiding a missile towards a moving target, said system comprising a detection device according to the invention.

[0030] The invention also relates to a computer program product comprising code instructions for implementing a method according to the invention, when the latter is executed by the processing unit of the detection device according to the invention.

[0031] Thus, the invention allows a comparison between the measured signals from the quadrants of the photodetector and levels which respect the theoretical energy distribution in order to validate the deviation measurements.

[0032] The invention makes it possible to identify erroneous deviation measurements caused by degraded operation.

[0033] The invention makes it possible to increase the reliability of the deviation function by adding a criterion on the geometries and radiometry of the spot on the photodetector.

[0034] Thanks to the invention, an analysis of the coherence of the signals with respect to the theoretical distribution (typically a disk) makes it possible to determine that the uniformity of the illumination in the optical pupil is significantly degraded following scintillation phenomena on the paths of the laser from the designator to the target and / or from the target to the sensor or even degradation due to the electronic chain as indicated above.

[0035] Knowing how to identify degraded operation, it is possible to implement corrective measures: invalidate aberrant deviations; temporally filter noisy deviations before their use by the tracking turret of the designation system (case of the LST sensor) or by the weapon guidance chain (case of the E-SAL sensor); if necessary, implement a device to reduce the effects of scintillation; being penalizing for optical transmission and therefore sensitivity, it is not desirable to have such a reducing device permanently in the optical channel; if necessary, implement a more robust adjustment of the electronic detection chains with respect to modulations under strong luminosity; being penalizing for sensitivity, it is not desirable to have to apply this adjustment systematically. PRESENTATION OF FIGURES

[0036] Other characteristics, aims and advantages of the invention will emerge from the following description, which is purely illustrative and non-limiting, and which must be read in conjunction with the appended drawings in which: there figure 1 illustrates a device for locating a target according to an embodiment of the invention; the figure 2 schematically illustrates a 4-quadrant photodetector; the Figures 3a and 3b illustrate detection configurations by a 4-quadrant photodetector; the figure 4 illustrates a degraded detection configuration using a 4-quadrant detector; the Figure 5 illustrates a sixteen-quadrant photodetector arranged in a matrix; the figure 6 illustrates steps of a method for locating a target according to an embodiment of the invention; the figure 7 illustrates a mode of selecting a configuration of this detector in a 4x4-quadrant matrix for a corresponding deviation measurement; figure 8illustrates several possibilities of 2x2 quadrant configuration selection for a corresponding deviation measurement of the figure 6 ; there figure 9 illustrates another example of distribution of a spot on a 4x4-quadrant detector; the figure 10 illustrates several possibilities for selecting a matrix with more than 4 quadrants for a corresponding deviation measurement of the figure 9 ; and the Figures 11a and 11b illustrate different quadrant arrangements to which the invention can be applied.

[0037] In all figures, similar elements have identical references. DETAILED DESCRIPTION Location device

[0038] The figure 1 illustrates a device 1 for locating a target T comprising a photodetector 2 with several quadrants and more precisely with NxM quadrants with N and M greater than or equal to two.

[0039] This 2-plane photodetector is behind an optic 6 through which light rays from the target T to be located pass. Such an optic 6 is for example a converging lens. In a complementary manner, the location device comprises a unit 3 for emitting a beam, known as a designator, making it possible to illuminate the target T. In the case of semi-active laser guidance, the emitted beam is a laser beam. In another example, the beam is emitted by a transmitter of a beacon carried by a remote-controlled munition, of the LED transmitter type or any other photonic transmitter associated with temporal modulation.

[0040] In the following we consider the case where the target returns an optical flow resulting from the illumination by the designator 3.

[0041] The location device 1 and the photodetector 2 are positioned relative to a sighting axis AA which forms with the direction BB of the target T a deviation θ. This deviation measurement θ is measured by means of the distribution of optical signals, i.e. the optical flux backscattered by the target, on the quadrants of the photodetector 2.

[0042] There figure 2 illustrates a four-quadrant photodetector, labeled A, B, C, D. In this example, the quadrants are in the form of an angular sector of a concentric crown, arranged in a matrix NxM=2x2 quadrants.

[0043] In operation, the photodetector 2 being centered on the line of sight AA, the optical signals detected by the photodetector 2 form a spot 21 centered as illustrated in the Figure 3a when the AA line of sight is aligned with the direction of the target; it is otherwise uncentered as shown in the Figure 3b . The center of the spot constitutes the coordinates of the deviation measurement θin site and in bearing and corresponds to the energy barycenter of the spot defined from the weighings according to the site and bearing axes. Each weighing is a function of the signal levels detected by each quadrant.

[0044] The photodetector 2 is configured to detect optical signals scattered by the target. It is connected to a processing unit 4 which in particular makes it possible to measure the deviation and to validate this measurement according to a method which will be described below.

[0045] The processing unit 4 comprises an electronic chain (not described or detailed because it is well known to those skilled in the art) connected to each quadrant of the photodetector 2 to acquire the electronic signals corresponding to the optical flow detected by each quadrant and integrated by the photodetection surface of the quadrant.

[0046] As indicated in the introduction, the deviation measurement can be distorted when the photodetector operates in degraded mode. figure 4illustrates a case of degraded operation, corresponding to the example of illumination of the Figure 3a : on the Figure 3a the energy distribution in the quadrants is nominal: the spot is present in its entirety and here the levels of the signals measured in each quadrant are equal. On the figure 4 , for the same deviation the energy distribution of spot 22 in the quadrants is inconsistent: the level of the signal measured on quadrant B is zero following a malfunction of quadrant B for example in the case of saturation of the electronic chain of quadrant B. There is an inconsistency between the distribution of the signals in the quadrants with the theoretical distribution associated with a deviation as calculated from the signals acquired in the case of figure 4 .

[0047] It is on the observation of this inconsistency that the method of the invention is based, a method implemented in the processing unit 4 of the device described above.

[0048] Such a principle applies to any photodetector 2 comprising several quadrants, arranged in a matrix of size NxM quadrants, NxM greater than or equal to four. Examples of quadrants conventionally arranged in a matrix are shown here, but this arrangement can be arbitrary, as illustrated as an example in the Figures 11a and 11b . Such arrangements other than a classic matrix can make it possible to minimize the number of processing channels to cover a field of view. This is for example the case of a 4x3 arrangement of quadrants corresponding to sectors of a concentric crown as illustrated in the Figure 11a .

[0049] Other arrangements may advantageously have a geometry that facilitates the connection of output signals by having each quadrant close to the periphery. This would be the case, for example, of the arrangement of 4x4 quadrants made up of triangles illustrated in the Figure 11b . On the Figures 11a and 11b the detected spot 23 is represented by a disk.

[0050] There Figure 5 illustrates an example of a photodetector 2 comprising a matrix of NxM=4x4 quadrants A 1 , A 2 , A 3 , A 4 , B 1 , B 2 , B 3 , B 4 , C 1 , C 2 , C 3 , C 4 , D 1 , D 2 , D 3 , D 4 . Each quadrant is configured to detect optical signals as explained previously. Process

[0051] In the context of a localization method described in relation to the figure 6electronic signals corresponding to the optical signals scattered by the target and detected by the photodetector 2 are acquired (step E1). We consider the case where the target is illuminated by a designator 3.

[0052] The acquired signals correspond to the optical signals diffused by the target T to be located and detected by the photodetector 2. One signal per quadrant is acquired.

[0053] These signals are then processed to determine a deviation in elevation and in bearing, noted respectively S(t) and G(t) (step E2). This deviation is directly linked to the position of the spot on the photodetector. In particular, the deviation is obtained from a calculation of weights which corresponds to comparisons of signal levels in different quadrants.

[0054] The measured deviation is obtained from the signals detected by each quadrant, as further detailed below.

[0055] Then, the measured deviation (step E2) is processed (step E4) to deduce indicators of the corresponding theoretical energy distribution of the optical signals, i.e. of the light spot, in the quadrants. These indicators are simply referred to as theoretical indicators.

[0056] In parallel, the acquired signals are processed (step E3) to calculate indicators of the corresponding energy distribution of the optical signals detected by the quadrants. These indicators are simply referred to as measured indicators.

[0057] We therefore have on the one hand indicators of the theoretical distribution and on the other hand indicators of the measured distribution.

[0058] These theoretical and measured indicators are then compared (step E5) to check their consistency and validate or not the measured deviation. Determination of the deviation in site S(t) and in bearing G(t) has from the acquired signals (step E2)

[0059] The determination of the deviation in site S(t) and in bearing G(t) includes a calculation (step E22) of the weighings according to these directions. These weighings exploit the signal levels detected in each quadrant and compare the signals on either side of the rows and columns of the quadrant matrix in the directions according to the site S(t) and bearing G(t) axes. In other words, the weighings compare the signal levels of several quadrants.

[0060] According to one embodiment, a four-quadrant detector is considered as in the figure 2 . We therefore have a matrix of 2x2 quadrants.

[0061] The deviation measurement is obtained from the on-site and on-bearing weighings. In the case of a 2x2 quadrant matrix, the on-site weighing is obtained by comparing two by two the signal levels detected by each quadrant along the on-site and on-bearing axes. In particular, the on-site weighing is given by PY = (SB + SD - SA - SC ) / (SA + SB + SC + SD ): it compares the signals on either side of the on-bearing axis (vertical) and the on-bearing weighing is given by PZ = (SA + SB - SC - SD ) / (SA + SB + SC + SD ): it compares the signals on either side of the on-site axis (horizontal).

[0062] According to one embodiment, in the case of an NxM quadrant photodetector with NxM greater than four as in the Figure 5, we consider subsets of 2x2 quadrants to simplify the calculations. The weighings in site and in bearing are considered by comparing two by two each quadrant around each row and column of the matrix of the subset of 2x2 quadrants in directions parallel to the site and bearing axes.

[0063] It has been illustrated on the figures 7 and 8 the choice of a four-quadrant configuration when spot 23 occupies four quadrants only out of the 4x4 quadrants of the detector.

[0064] On the figure 7 task 23 occupies four quadrants: A 1 , A 2 , B 1 , B 2 . This configuration offers several possibilities for choosing a subset of 2x2 quadrants. The figure 8illustrates these possibilities C HG , C HC , C HD , C CG , C CC , C CD , C BG , C BC , C BD . For the calculation of the deviation we must choose one of these configurations among those which detect signals. These are the configurations C HG , C HC , C CG and C CC since they have at least one quadrant which detects the signal. However, we note that only the configuration C HG detects the signal on its four quadrants.

[0065] Therefore, the method comprises a sub-step of selecting (step E21) a 2x2 quadrant configuration for calculating the weights, the selected configuration being the one for which the detected signal level is the highest. In other words, it is the one which will give the highest signal-to-noise ratio which will be retained for the rest of the method.

[0066] Returning to the example of the figure 7 ,the deviation is thus calculated on the signals from quadrants A 1 , A 2 , B 1 and B 2 . It is understood that the weighings PY and P z have expressions identical to those given above, but applied to the selected 2x2 quadrant configuration. That is to say that in this embodiment, we return to the four-quadrant situation illustrated above in connection with the Figure 3a .

[0067] According to another embodiment, a situation is considered in which the spot 23 occupies more than four quadrants on the NXM detector quadrants. In this case, to improve the detection, it is appropriate to consider more than 2x2 quadrants. This is what the figures 9 And 10 .

[0068] On the figure 9, spot 23 reaches eight contiguous quadrants which are inscribed in the example in a 3x3 quadrant matrix covering: A 1 , A 2 , A 3 , B 1 , B 2 , B 3 , C 1 , C 2 , C 3 . This sub-matrix is ​​delimited by the dotted lines on the figure 9 . This is the smallest matrix that includes all the quadrants reached by task 23. Thus, when task 23 covers more than four quadrants, the weights are established according to a principle similar to the 4-quadrant cases but which compares more weights for example: comparison of: A 1 -B 1; A 2 -B 2 ; A 3 -B 3 ; (A 1 +A 2 +A 3 ) - (B 1 +B 2 +B 3 ) ; comparison of: B 1 -C 1; B 2 -C 2 ; B 3 -C 3 ; (B 1 +B 2 +B 3 ) - (C 1 +C 2 +C 3 ); comparison of: A 1 -A 2; B 1 -B 2 ; C 1 -C 2 ; (A 1 +B 1 +C 1 ) - (A 2 +B 2 +C 2 ) ; comparison of: A 2 -A 3; B 2 -B 3 ; C 2 -C 3 ; (A 2 +B 2 +C 2 ) - (A 3 +B 3 +C 3 ).

[0069] More generally, the quadrants detecting a signal define a NixMi matrix of quadrants included in the NxM matrix of quadrants of the photodetector, Ni less than or equal to N, Mi less than or equal to M, and the weighings in site and bearing exploit the signals detected in this matrix of NixMi quadrants detecting signals, the weighings in site and bearing being respectively obtained by comparing the signals detected on adjacent quadrants taken along the rows of this NixMi matrix on the one hand and taken along the columns of this NixMi matrix on the other hand. The NixMi matrix is ​​preferably the smallest matrix containing the quadrants reached by the spot 23, that is to say detecting a signal.

[0070] Thus, the NixMi matrix of quadrants detecting a signal is scanned to compare the signals according to rows and columns.

[0071] In this case, sub-step E21 described above is not implemented and the weight calculation step E22 takes into account more than 2x2 quadrants.

[0072] In parallel with the calculation of the weights, a total signal-to-noise ratio RSB tot is calculated from the detected signals (step E23). Such a signal-to-noise ratio makes it possible to adapt the criterion used to check the consistency of the distribution criteria with the dispersions on the signal levels in the quadrants taking into account this signal-to-noise ratio. Here, we adapt to the “quality of the measurement”.

[0073] Then, from deviation tables Tab1 stored in the location device 1, an interpolation is implemented to associate with each weighing measurement, the deviation values ​​G(t) and S(t) in bearing and in site respectively (step E24). In an implementation of the invention, a bilinear interpolation is applied, which has the advantage of being simple to implement.

[0074] It is understood that the deviation tables depend directly on the geometry of the photodetector.

[0075] The deviation tables are produced in the factory during calibration of the photodetector on a discretization of its angular operating range. For each point in the range, the signals output from the quadrants are recorded. The "laboratory" environment of this factory characterization guarantees the absence of disturbance of the laser signals and therefore makes it possible to constitute a model of the nominal reference operation.

[0076] This factory characterization allows the construction of a theoretical map, consisting of the expected reference values ​​to calculate a metric at any point in the field. This metric is used to verify the consistency of the measurements. This mapping dedicated to "validity" is added to the usual characterization necessary for the needs of the "deviation measurement" functionality, which allows the establishment of correspondences between weighings and deviation measurements. The "validity" map can therefore use the same source data and be created from the same factory test protocol.

[0077] To create these maps, we generally proceed as follows: a collimated laser flux generated by a bench is directed towards the photodetector according to controlled sites and deposits; for each orientation of the incident laser flux, the signals on each of the quadrants of the photodetector are measured; the theoretical correspondence table between weighings and deviation measurements is constructed.

[0078] At the end of this step E2, the deviations in bearing G(t) and in site S(t) are obtained. These deviations are used to locate the target T in relation to the line of sight of the device if they are validated in the subsequent steps. This deviation is also used to track a target once located. Determination of indicators of signal distribution detected (step E3)

[0079] From the acquisition of electronic signals on each of the channels (quadrant) of the photodetector, indicators of the distribution of the detected signals are determined here (step E31).

[0080] The measured distribution indicators include, in addition to the weights used for deviation measurement, half-weights. Half-weights are also weights but only take signals two by two.

[0081] In the case of four quadrants, PY and PZ are the weights which compare respectively the top and bottom signals, and in the case of four quadrants P Y1 , P Y2 , P Z1 , P Z2 are strictly speaking half-weights since these weights compare the signals two by two, for P Y1 , the two top signals, for P Y2 the two bottom signals, for P Z1 the two left signals, for P Z2 the two right signals and are given by the following expressions: P Y1 =(SB -SA ) / (SA +SB ), weighing comparing the two signals at the top; P Y2 =(SD -SC ) / (SC +SD ), weighing comparing the two signals at the bottom; PY =(SB +SD -SA -SC ) / (SA +SB +SC +SD ) weighing comparing the signals on either side of the vertical axis; P Z1 =(SA -SC ) / (SA+SC ), weighing comparing the two signals on the left; P Z2 =(SB -SD ) / (SB +SD ), weighing comparing the two signals on the right; PZ =(SA +SB -SC -SD ) / (SA +SB +SC +SD ) weighing comparing the signals on either side of the horizontal axis.

[0082] By adding the calculation of half-weights to that of the weights used to estimate the deviation, we create redundancy in the measured information.

[0083] Indeed, a deviation (site or bearing) is associated with a single weighing value (bijectivity), while the half-weights can take several pairs of values ​​depending on the deviation on the other axis (bearing or site respectively). However, only a pair of values ​​for the half-weights verifies the geometric and radiometric coherence expected in theory for the distribution of light energy on the different quadrants associated with this position in site and bearing of the target.

[0084] These calculations are transposable to any multi-quadrant configurations.

[0085] In the case where there are more than four quadrants, the idea is to calculate, in addition to the weights necessary for the deviation measurement, other quantities associated with the distribution of the spot 23 on the photodetector. Thus, the person skilled in the art understands that a multitude of indicators can be calculated.

[0086] From these weighings, a metric Ω is then determined (step E32) from the measured indicators. This metric Ω is intended to characterize the energy distribution of the laser spot 23 on the different quadrants of the photodetector. In one example, an easily applicable metric Ω is based on an unsigned maximum deviation between the weighings and the half-weighings. Such a metric Ω is for example given by the following expression in the case of four quadrants:

[0087] Of course, in the case of configurations with more than four quadrants, it is always a matter of calculating a metric based on the weights and half-weights.

[0088] Other metrics could be used, for example a metric based on distance calculations with reference to a spot that respects the theoretical shape. Determination of indicators of the theoretical distribution of signals for the acquired deviation (step E4)

[0089] From the deviations measured in step E2, by interpolation, for example bilinear interpolation, with theoretical maps of weights and half-weights Tab2, theoretical indicators are determined for this measured deviation (step E41).

[0090] Here it is a question of obtaining the expected weights and half-weights for the deviation measured by symmetry to obtain the theoretical indicators of the distribution of the acquired signals.

[0091] These indicators are noted here P' Y , P' Z , P' Y1 , P' Y2 , P' Z1 , P' Z2 .

[0092] The maps of the theoretical indicators Tab2 (here theoretical weights and half-weights) are obtained in the factory in a similar way to the maps Tab1.

[0093] These theoretical indicators have the same expressions as the measured indicators except that they were calculated from the electronic signals S' A , S' B , S' C , S' D measured in the factory on the quadrants.

[0094] We can then characterize the energy distribution of the laser spot on the quadrants of the photodetector corresponding to these theoretical indicators (step E42). We apply a metric Ω' which has the same definition as the metric Ω applied in step E32.

[0095] If we take the previous example, the metric Ω' is thus based on an unsigned maximum difference between the theoretical weights and the theoretical half-weights, and given by the following function: Verification of the consistency of the indicators of the theoretical distribution and the measured distribution (step E5)

[0096] The verification of the consistency between the indicators of the theoretical and measured distributions is advantageously implemented by comparison (step E51) of the metrics Ω and Ω'' and includes the calculation of a comparison criterion between these metrics.

[0097] Such a criterion is given by the following formula: Crit = abs(Ω - Ω'') namely an unsigned difference of the two metrics.

[0098] This criterion is then compared to a threshold (step E52).

[0099] There are two cases: first case: if Crit ≥ S then the current deviation is invalid. In other words, if the unsigned difference between the two metrics Ω' and Ω is greater than or equal to the threshold value, then this deviation is considered "aberrant or invalid"; second case: if Crit

[0100] According to one embodiment, the threshold S is set arbitrarily and advantageously results from a compromise between a false alarm rate characterizing the proportion of valid deviations that are invalidated; a detection rate characterizing the proportion of aberrant deviations identified as such by the criterion. ​

[0101] In addition (step E53), it may be necessary, depending on the compromise sought, to add one or more threshold dependencies such as a dependence on the signal-to-noise ratio RSBtot of the measured signals, which makes it possible to stabilize the rate of “false alarms” over the operating domain by an increase in the threshold value at low RSBtot and by a decrease in this value at high RSBtot, a dependence on the position in the angular domain, which makes it possible to adjust the performance of the coherence calculation in the field.

[0102] In the first case, the signal-to-noise ratio RSBtot of the measured signals is used to define the applicable threshold, by interpolation of Tab3 data established for some angular positions in site and bearing, and functions of some tabulated levels of the RSBtot. From then on the threshold depends on the measured deviation.

[0103] In the second case, an interpolation is performed in a threshold map according to the measured (current) deviation using Tab3 data.

[0104] Following the comparison with the threshold, a validity status of the measured deviation is obtained. Other applications

[0105] The method has been described in the context of a photodetector having a matrix of several quadrants.

[0106] However, different variants are possible and they result from the same principle which consists of exploiting the a priori knowledge of the energy distribution of the spot on a multi-quadrant photodetector.

[0107] The spot may have a geometry other than a uniform disk. Other shapes are sometimes preferred, to simplify or compact the optical element upstream of the photodetector and / or as a result of integration into a device combining several optronic channels, for example in the case of equipment (with central shutter; and / or to favor certain specifications of the deviation response, for example, a reduced dependence on the size of the spot on the target and / or on the distance between the target and the SAL sensor, whether it is of the LST or E-SAL type.

[0108] These other spot geometries on the photodetector may in particular be: a crown-type shape, a “Gaussian” type shape, a “square” type shape, a shape among those mentioned above and presenting a spatial modulation resulting from optical elements for reducing scintillation effects.

[0109] The multi-quadrant photodetector can be of the 4-quadrant type as more particularly taken as an example above, this is the usual geometry; but it can have any other geometry which, associated with the theoretical size and shape of the laser spot, makes it possible to identify an anomaly in the distribution of energy on the quadrants. In particular, as presented above, a first family of extensions covers photodetectors with N x M elements, each element being square or rectangular. Another family of extensions groups together photodetectors with geometries having a number of angular sectors greater than 4.

[0110] Laser deviation measurement using a multi-element photodetector is applied in many fields.

[0111] In the context of semi-active weapon guidance or laser target pointing, a target intercepts the beam and scatters some of the energy towards the SAL sensor.

[0112] In other contexts, multi-element photodetectors are associated with lasers in direct configurations. In these direct configurations, the laser is directed towards the sensor equipped with a multi-element photodetector and the sensor develops a deviation measurement to assist in the pointing and / or relative movement of a module carrying the laser and a module carrying the sensor. This type of direct configuration instrumentation is used in multiple fields, such as the space field, the field of industrial machines, the field of public works and the field of autonomous vehicle guidance, particularly in cooperative contexts. In these last two fields, one can encounter the disturbing phenomena to which the invention provides a response: the problems of non-homogeneity caused by atmospheric scintillation, and the problems of parasitic fluxes.

Claims

1. A method for locating a target (T) by means of a detecting device comprising a photodetector comprising NxM quadrants positioned with respect to a sighting axis of the detecting device, the photodetector being configured to detect optical signals from the target to be detected, the photodetector comprising quadrants distributed along two perpendicular axes called elevation Z and azimuth Y axes, said method being implemented in a processing unit of said device and comprising the following steps, the target scattering an optical signal: - acquiring (E1) the signals detected by each quadrant, which correspond to the signals (SA, SB, SC, SD) scattered by the target toward the quadrants; - processing (E2) the acquired signals to determine an elevation angle error and an azimuth angle error of the direction of the target with respect to the sighting axis of the device; - processing (E3) the acquired signals with a view to deducing therefrom indicators of the distribution of the signals detected by the quadrants; - processing (E4) the determined elevation angle error and azimuth angle error with a view to deducing therefrom indicators of the theoretical distribution of the signals in the quadrants for these angle errors; distribution indicators (E3, E4) of the signal in the quadrants being obtained by means of theoretical and measured weighting values, said weightings being based on the levels of the signals detected by the quadrants, - verifying (E5) the coherence of the theoretical distribution indicators with the distribution indicators obtained using the acquired signals so as to determine whether the measured angle error is valid.

2. The method according to claim 1, wherein, the theoretical distribution indicators of the signals (E4) in the quadrants is obtained by interpolation, preferably bilinear interpolation, of the angle errors with a mapping of the theoretical distribution indicators of the signals as a function of a plurality of theoretical angle errors.

3. The method according to one of claims 1 to 2, wherein the NxM quadrants of the photodetector are distributed along two mutually perpendicular axes, referred to as the elevation axis Z and azimuth axis Y, the quadrants detecting a signal defining a sub-matrix included in the NxM matrix, of size NixMi quadrants, with Ni less than or equal to N, and Mi less than or equal to M, and detecting signals, the distribution indicators of the signals being obtained by comparing the signals detected on adjacent quadrants taken along the rows of the NixMi matrix on the one hand, and taken along the columns of the NixMi matrix, on the other hand.

4. The method according to one of the preceding claims, wherein verifying the coherence consists of calculating a comparison criterion between a first metric based on the theoretical distribution indicators and a second metric based on the distribution indicators of the acquired signals, the criterion being compared with a threshold in order to validate the angle error measurements.

5. The method according to claim 4, wherein the comparison criterion is defined by an absolute difference between the first and second metrics, the angle error being valid if this criterion is less than a threshold and invalid if this criterion is greater than or equal to said threshold.

6. The method according to one of claims 4 to 5, wherein the threshold is adjusted as a function of the signal-to-noise ratio measured on each quadrant and / or of the determined angle error.

7. The method according to one of the preceding claims, wherein the photodetector comprises four quadrants A, B, C, D, quadrants A and B being above the elevation axis Z, quadrants C and D being below the axis Z, quadrant C being below quadrant A, quadrant D being below quadrant B, the distribution criteria of the signals over the quadrants being defined in the following manner: - Y1=(B-A) / ( A+B), weighting comparing the top two signals; - Y2=(D-C) / (C+D), weighting comparing the bottom two signals - Y=(B+D-A-C) / (A+B+C+D) weighting comparing the signals on either side of the vertical axis; - Z1=(A-C) / (A+C), weighting comparing the two signals on the left - Z2=(B-D) / (B+D), weighting comparing the two signals on the right - Z=(A+B-C-D) / (A+B+C+D) weighting comparing the signals on either side of the horizontal axis.

8. The method according to claim 7, wherein the metrics based on theoretical or measured distribution criteria are defined by max([abs(Y1-Y2); abs(Z1-Z2); abs(Y1-Y); abs(Z1-Z); abs(Y2-Y), abs(Z2-Z)]) with max the maximum value and abs the absolute value.

9. A device for detecting a target comprising a photodetector with a plurality of quadrants and a processing unit configured to implement a method according to one of the preceding claims.

10. A system for guiding a missile towards a moving target, said system comprising a detection device according to the preceding claim.

11. A computer program product comprising code instructions for implementing a method according to one of claims 1 to 8, when it is executed by a processing unit of a device according to claim 9.